SOLID-PARTICLE ABRASION OF HYDRAULIC CONCRETE
Yu Wen Liu(1), Huang Hsing Pan(2), and Chao-Shun Chang (3)
(1) Dept. of Civil and Water Resources Eng. National Chiayi University, Taiwan
(2) Dept. of Civil Engineering, Kaohsiung University of Applied Sciences, Taiwan
(3) Dept. of Construction Eng. National Kaohsiung First University of Science and
Technology, Taiwan
Abstract ID Number (given by the scientific editors/organizers): …124..
Keywords: Abrasion, Hydraulic Concrete, Impacting, erodent size
Author contacts
Authors
E-Mail
Fax
Postal address
-886-5-2717693 300 University road,
Chia Yi 600, Taiwan
415 Chien-Kung Road,
Huang Hsing Pan [email protected]
-886-7-3962074 Kaohsiung 807, Taiwan
1 University Road,
Chao-Shun Chang [email protected] -886-7-6011017 Yenchao, Kaohsiung,
824 Taiwan
Contact person for the paper: Yu Wen Liu
Yu Wen Liu
[email protected]
Presenter of the paper during the Conference: Yu Wen Liu
9
Total number of pages of the paper (the first pages and
the licence to publish excluded):
8
SOLID-PARTICLE ABRASION OF HYDRAULIC CONCRETE
Yu Wen Liu(1), Huang Hsing Pan(2), and Chao-Shun Chang(3)
(1) Dept. of Civil and Water Resources Eng. National Chiayi University, Taiwan
(2) Dept. of Civil Engineering, Kaohsiung University of Applied Sciences, Taiwan
(3) Dept. of Construction Eng. National Kaohsiung First University of Science and
Technology, Taiwan
Abstract
Most of the abrasion damage is caused by the action of water-borne particles (silt, sand,
gravel, and other solid) impacting and rolling against the concrete surface during hydraulic
structure operation. In this paper solid-particle abrasion of concrete containing slag was
studied. Experiments included use of angular river sand abrade of mean diameter ~ 0.6, 1.2,
2.5 and 5 mm impacting at 30°, 45° and 90° to the concrete surface. The waterborne sand
flow impact test method was used. Test results show that the abrasion rate to be a strong
function of erodent size. As the erodent size increased from 0.6 mm to 1.2mm, 2.5 mm, then
to 5mm, the abrasion rate of concrete increased from 123% to 262% and 385%. Moreover, the
abrasion rate of concrete impacted at 90° was higher than of that of 30° and 45°. Material loss
in concrete appears to have been caused by a complex combination of fracture mechanisms.
Keywords: Abrasion, Hydraulic Concrete, Impacting, erodent size
1. INTERDICTION
In Taiwan, all rivers originate from the peak of each ridge, snaking through valleys and
running across sporadic plains to reach the ocean. Because of high ridge peaks and steep
valley basins, all rivers are short and steep causing rapid flow during storms, particularly
during the typhoon season. There is a high average annual rainfall of 2530 mm in Taiwan,
Page 1
approximate 2.6 times of world average rainfall. In addition, the type and space distribute of
rainfall do not exceed each other much. The rainfall is concentrated in the month of May to
October, where approximately 78 % of the average annual rainfall occurs [1]. Furthermore,
because of the country’s frequent earthquakes and fragile geology, the rapid flow of rivers
carry heavy sand and gravel, making the sediment yield per area and sand percentage of river
given more than ten times that of the world average. As a result, the most significant abrasion
problems happen due to the abrasion effect of the friction and impact of waterborne sand on
the hydraulic structures concrete surface.
When a brittle material is impacted by a hard sharp particle, the contact area is plastically
deformed due to the high compressive and shear stresses and a radial crack is formed. After
the impact, the plastic deformation leads to large tensile stresses that resulted in lateral cracks
causing the material removal [2-3]. Abrasion condition and abrade characteristics also play
key roles in determining abrasion rate. Large, hard particles are expected to import maximum
abrasion rate. Large abrade particles flow much better than small one, and the debris that
forms with import by larger abrades is larger [4]. Therefore, in this paper the waterborne sand
flow test which combining the water-jet impact load and sand particle shear/friction forces
produced by a hydro-particle flow, was used to investigated the effect of impact angle, abrade
particles size on abrasion resistance of hydraulic concrete.
2. EXPERIMENTAL PROGRAM
2.1 Materials
Materials used in manufacturing test slabs include: (1) Type I Portland cement (ASTM
C150); (2) river sand having a fineness of 2.95, a specific gravity of 2.64, and an absorption
of 1.2 %; (3) crushed basalt coarse aggregate with a maximum aggregate size (Dmax) of 13
mm, specific gravity of 2.64, absorption of 1.0 %, and dry-rodded density of 1665 kg/m3, and
(4) slag furnace with a specific gravity of 2.89 supplied by China Hi-Ment corporation; (5)
superplasticizer (SP) conforming to ASTM C494 Type-G with a specific gravity of 1.1; and (6)
fresh water.
2.2 Mixture proportions
The mixture proportions used in this investigation were designed to study the effect of
abrasion type on concrete using the weight and absolute-volume method. As summarized in
Table 1, for concrete mixtures were prepared with three different water-to-cementitious
material ratios (w/cm) of 0.28 0.36 and 0.50. The cement was partially replaced with 20% of
slag furnace by weight. A superplasticizer was used to produce concrete having roughly the
Page 2
same slump of 22±2cm. The compressive strength of concrete was shown in Table 1.
Table 1 - Concrete mix proportions and Compressive strength
Batch w/cm
SP
Slump
(cm)
Compressive
strength
(MPa)
Dry quantities, kg/m3
Water Cement
SF
Sand Gravel
C28
0.28
160
457
114
730
925
12.5
24
90.8
C36
0.36
160
356
89
780
985
10.9
22
50.3
C50
0.50
160
256
64
820
1070
0.5
21
30.4
2.3 Experimental method
The abrasion tests were carried out in a waterborne sand flow apparatus that is described in
Refs. [5-6]. The abrade particles were angular quartz tic river sand with a Mohs-hardness (Hp)
of 8, density (ρp) of 2.64 g/㎝ 3 and mean diameters (D) of 0.6, 1.2, 2.5 and 5 mm, respectively.
The waterborne sand flow velocity was 10±1 m/s and the angle of impact was 30°, 45° or 90°.
The abrasion rate (AR, in g/h) was determined from the specimen mass loss vs. test time. A
minimum of three measurements were used to establish AR. The values of AR obtained are
estimated to be accurate to ±15%. The larger error quoted here is attributable primarily to
specimen uniformity, not reproducibility of the testing procedure.
3. RESULTS AND DISCUSSION
Observations on the specimen after being subjected to a waterborne sand jet test reveal that
transient hydraulic rim-pulls impinged on the specimen and caused local tensile stresses in the
top layer of the exposed concrete. Based on the energy conservation theory, the intensity of
the tensile stresses varied in respect to the impact momentum of the hydraulic jet forces.
These tensile stresses are the prime culprits for causing cracks in the hardened mortar and
fractures around aggregate particles which eventually lead to impact abrasion.
Fig. 1 shows photos from various impact abrasions of the concrete after testing. The matrix
exhibits significant indenting by the exposed erodent, the aggregate grain appears to peel
away and the mortar on which interfacial cracks become visible on the concrete prepared with
high w/cm and impacted at 90° (Fig. 1a), whereas it appears to be rather smooth in low w/cm
concrete and impacted at 45° (Fig. 1b). The SEM revealed the cracks formed in the cement
matrix and the interface between aggregate grain and cement matrix shown in Fig. 2a for
concrete impacted at 90°. The cement matrix significant indenting by the abrodent, with
Page 3
concomitant smearing of the surface, and formation of many small cracks rather than a few
large ones. In addition, the concrete impacted at 90° displayed a rougher and more rugged
surface than concrete impacted at 45° and 30° (Fig. 2b, 2c).
10 mm
10 mm
(a) w/cm = 0.50, impacted at 90°
(b) w/cm = 0.28, impacted at 45°
Fig. 1 Images of worn concrete surfaces under various conditions
Fig. 3 shows the relationship between impact angle and abrasion rate. From Fig. 3 it is
evident that the impact angle of 90° has the highest abrasion rate, and the impact angle of 45°
has the smallest one. For concrete made with w/cm of 0.36 and 0.50, and impacted at 90°, the
abrasion rate was nearly 124 % and 207 % higher than that of 45°, respectively.
A fundamental approach was to obtain the brittle abrasion deals with material removal due
to crack formation, while ductile abrasion deals with material removal due to cutting and
plowing [7]. For concrete, it is generally considered that abrasion damage is the gradual
removal of material caused by repeated deformation and cutting action. The theoretical
analysis for cutting [8-9] shows that progressive cutting occurs at a given low impact angle,
under which a particle may slip on a surface or it may retain some of its own impact energy
after the impact, resulting in a decreased in material removal. Moreover, the abrasion rate is
associated with the relation between the shear force to cut a mass of material and the material
resistance indicated by the compressive strength or hardness. For the concrete impacted by
waterborne sand flow the abrasion action mainly include pre-abrasion peeling by water
molecules and its associated hydraulic pressure, solid particle impact, edge effect and prising.
For the concrete specimen impacted at 90° the crack formation due to normal component of
impact velocity dominated material removal, while impacted at 30° the cutting dominated
material removal. For waterborne sand flow test, it can be found that the abrasive force due to
Page 4
(a)
(b)
(c)
Fig. 2 SEM-images of worn concrete surface, scale: 60 μm. (a)
impacted at 90°, (b) impacted at 45° and (c) impacted at 30°
Page 5
normal component of impact velocity is higher than the cutting. With impact at 45°, SEM
revealed that the indentation of the surface was insignificant compared with impact at 30° and
90°, reducing less material loss.
On the other hand, the abrasion rate is a strong function of erodent size as shown in Fig. 4.
As the erodent size increased from 0.6 mm to 1.2mm, 2.5 mm, then to 5mm, the abrasion rate
of concrete increased from 123% to 262% and 385%. When abrade size is decrease,
eventually the abrade particles are not able to initiate cracking and will only plastically
deform the target. Theories of abrasion of brittle materials, which are based on elastic-plastic
interactions [10], predict AR α D2/3. The experimental data for concrete specimens show a
distinct relationship between the abrasion rate and the abrade size. It can be approximated by
a non-linear regression of AR α D3/2 with a regression coefficient of R2=0.912. The deviations
from the ideal are common and are usually related to interfacial, microstructural and flaw
effects. Moreover, the water flow due to establishment of a stagnation pressure enters
pre-existing flaws in the material, especially microcracks in the interfacial zones between
paste and aggregate.
50
W/B=0.50
W/B=0.36
Abrasion rate (g/h)
40
30
20
10
0
30°
45°
60°
90°
Impact angle
Fig. 3 Abrasion rate versus impact angle for concrete made with W/B= 0.5 and 0.36
Page 6
35
Abrasion rate (g/h)
30
25
20
15
1.5627
y = 1.8971x
2
R = 0.9122
10
5
0
0.6
1.2
2.5
5
Abrade average size (mm)
Fig. 4 Abrasion rate versus abrade average size for concrete made with W/B=0.36
4. CONCLUSSIONS
Solid-particle abrasion rate of concrete depended strongly on abrade size and impact angle.
The abrasion rate was highest at 90° impact and lowest at 45° impact. As the erodent size
increased from 0.6 mm to 1.2mm, 2.5 mm, then to 5mm, the abrasion rate of concrete
increased from 123% to 262% and 385%. It can be approximated by a non-linear regression
of AR α D3/2 with a regression coefficient of R2=0.912. For the concrete impacted by
waterborne sand flow the abrasion action mainly include pre-abrasion peeling by water
molecules and its associated hydraulic pressure, solid particle impact, edge effect and prising
Material loss in concrete appears to have been caused by a complex combination of fracture
mechanisms.
ACKNOWLEDGMENTS
This paper was supported by the Taiwan National Science Council under NSC
94-2211-E-415-003.
REFERENCES
[1] http://river.nchu.edu.tw, Soil and water preservation code, engineering, council of agriculture,
Taiwan.
[2] Henk Wensink, Miko C. Elwenspoek, A closer look at the ductile-brittle transition in solid particle
Page 7
erosion, Wear 253, (2002)1035-1043.
[3] Hocheng , H., and Weng, C.H., Hydraulic Erosion of Concrete by a Submerged Jet, J. of Materials
Eng. and Performance, 11(3) (2002) 256-261.
[4] K.C. Goretta, Solid-particle erosion of Portland cement and concrete, Wear 244 (1999) 106-112.
[5] Yu-Wen Liu, Tsong Yen, Tsao-Hua Hsu, Abrasion Erosion of Concrete by Waterborne Sand,
accepted Cement and Concrete Research.
[6] Y. W. Liu, Abrasion-Erosion Resistance of High Performance Concrete Used as Overlay Material
of Hydraulic Structures, PhD dissertation, Department of Civil Engineering University of
Chung-Hsiung, Taiwan, (2005).
[7] G.L. Sheldon, I. Finnie, On the ductile behavior of nominally brittle materials during erosion
cutting, J. Eng. Ind. 88 (1966) 387-392.
[8] A.A. Munoz, P. Sheng, An analytical approach for determining the environmental impact of
machining processes, J. Mater. Process. Technol. 55(3-4) (1995) 736-758.
[9] Y.I.Oka, H. Ohnogi, T. Hosokawa, M. Matsumura, The impact angle dependence of erosion
damage caused by solid particle impact, Wear203-204 (1997) 573-579.
[10] A.G. Evans, M.E. Gulden, M. Rosenblatt, Impact damage in brittle materials in the elastic-plastic
response regime, Proc. R. Soc. London 361(1706) (1978) 343-365.
Page 8